Clinical application has always been the final test of the suitability of dental materials. In recent years the number of clinical investigations of materials has increased dramatically, especially since the adoption of legislation requiring clinical evaluation and demonstration of suitability before new products can be marketed. This sudden increase in clinical evaluation has prompted a marked diversification in the methods used for the evaluation of restorations and analysis of the results. There has been little apparent attempt to determine whether the methods used were appropriate or effective. This study proposes to examine the methods used for the clinical evaluation of marginal breakdown in dental amalgams and occlusal wear in composite restorations, as well as the statistical methods used for the analysis of these measurements. In addition, it will investigate the effect of experimental design factors and operator-evaluator-materials interactions on the effectiveness of clinical investigations in restorative dentistry. The investigation will include both direct clinical evaluations and indirect evaluations using physical records of the restorations. It will take advantage of the extensive records accumulated in controlled clinical evaluations of materials in the Operative Dentistry Research Clinic over the last fifteen years. Initial emphasis will be placed on laboratory evaluations of physical records for comparison with previously recorded clinical evaluations. This will be supplemented by additional clinical evaluations to extend the evaluation period for selected restorations and to test new and proposed methods. All such evaluations will be combined with continuing active studies. The need of improved methods increases as the materials are gradually improved since improved materials require longer times or larger sample sizes for any given method to detect differences between materials and techniques. Identification of improved evaluation methods can result in significant savings in the time and effort needed to evaluate materials.